A comprehensive AI model development framework for consistent Gleason grading
-
Published:2024-05-09
Issue:1
Volume:4
Page:
-
ISSN:2730-664X
-
Container-title:Communications Medicine
-
language:en
-
Short-container-title:Commun Med
Author:
Huo XinmiORCID, Ong Kok HaurORCID, Lau Kah Weng, Gole Laurent, Young David M., Tan Char LooORCID, Zhu Xiaohui, Zhang Chongchong, Zhang Yonghui, Li Longjie, Han HaoORCID, Lu Haoda, Zhang Jing, Hou Jun, Zhao Huanfen, Gan Hualei, Yin Lijuan, Wang Xingxing, Chen Xiaoyue, Lv Hong, Cao Haotian, Yu Xiaozhen, Shi Yabin, Huang Ziling, Marini Gabriel, Xu JunORCID, Liu Bingxian, Chen Bingxian, Wang Qiang, Gui Kun, Shi Wenzhao, Sun Yingying, Chen Wanyuan, Cao Dalong, Sanders Stephan J.ORCID, Lee Hwee Kuan, Hue Susan Swee-ShanORCID, Yu WeimiaoORCID, Tan Soo YongORCID
Abstract
Abstract
Background
Artificial Intelligence(AI)-based solutions for Gleason grading hold promise for pathologists, while image quality inconsistency, continuous data integration needs, and limited generalizability hinder their adoption and scalability.
Methods
We present a comprehensive digital pathology workflow for AI-assisted Gleason grading. It incorporates A!MagQC (image quality control), A!HistoClouds (cloud-based annotation), Pathologist-AI Interaction (PAI) for continuous model improvement, Trained on Akoya-scanned images only, the model utilizes color augmentation and image appearance migration to address scanner variations. We evaluate it on Whole Slide Images (WSI) from another five scanners and conduct validations with pathologists to assess AI efficacy and PAI.
Results
Our model achieves an average F1 score of 0.80 on annotations and 0.71 Quadratic Weighted Kappa on WSIs for Akoya-scanned images. Applying our generalization solution increases the average F1 score for Gleason pattern detection from 0.73 to 0.88 on images from other scanners. The model accelerates Gleason scoring time by 43% while maintaining accuracy. Additionally, PAI improve annotation efficiency by 2.5 times and led to further improvements in model performance.
Conclusions
This pipeline represents a notable advancement in AI-assisted Gleason grading for improved consistency, accuracy, and efficiency. Unlike previous methods limited by scanner specificity, our model achieves outstanding performance across diverse scanners. This improvement paves the way for its seamless integration into clinical workflows.
Funder
Agency for Science, Technology and Research
Publisher
Springer Science and Business Media LLC
Reference38 articles.
1. Ferlay, J. et al. Cancer statistics for the year 2020: an overview. Int. J. Cancer 149, 778–789 (2021). 2. Di Cataldo, S. & Ficarra, E. Mining textural knowledge in biological images: applications, methods and trends. Comput. Struct. Biotechnol. J. 15, 56–67 (2017). 3. Bera, K., Schalper, K. A., Rimm, D. L., Velcheti, V. & Madabhushi, A. Artificial intelligence in digital pathology—new tools for diagnosis and precision oncology. Nat. Rev. Clin. Oncol. 16, 703–715 (2019). 4. Van der Laak, J., Litjens, G. & Ciompi, F. Deep learning in histopathology: the path to the clinic. Nat. Med. 27, 775–784 (2021). 5. Alom, M. Z. et al. A state-of-the-art survey on deep learning theory and architectures. Electronics 8, 292 (2019).
Cited by
1 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献
|
|